[ App ] Update Multi_Input example #2743
Conversation
- This commit is related to issue nnstreamer#2660 - When using multi-inputs, users must feed the data in reverse order due to a known bug that needs fixing. In the current version, the input must be provided in reverse order, which was not shown in the previous example where random data with the same dimensions were used. - To provide a more accurate example to NNTrainer users, I have temporarily updated this example. - Once the issue is handled, further updates will be necessary. Signed-off-by: Eunju Yang <[email protected]>
EunjuYang
requested review from
myungjoo,
jijoongmoon,
again4you,
jaeyun-jung,
leemgs,
wooksong,
helloahn,
kparichay,
gichan-jang,
anyj0527,
zhoonit,
lhs8928,
songgot,
jihochu,
DonghakPark,
SeoHyungjun,
baek2sm,
skykongkong8,
djeong20 and
a team
as code owners
|
📝 TAOS-CI Version: 1.5.20200925. Thank you for submitting PR #2743. Please a submit 1commit/1PR (one commit per one PR) policy to get comments quickly from reviewers. Your PR must pass all verificiation processes of cibot before starting a review process from reviewers. If you are new member to join this project, please read manuals in documentation folder and wiki page. In order to monitor a progress status of your PR in more detail, visit http://ci.nnstreamer.ai/. |
|
To contributor, We have used 'Signed-off-by:' notation by default to handle the license issues, that result from contributors. Note that 'Is there a Signed-off-by line?' is important because lawyers tell us we must have to it to cleanly maintain the open-source license issues even though it has nothing to do with the code itself. |
There was a problem hiding this comment.
@EunjuYang, 💯 All CI checkers are successfully verified. Thanks.
EunjuYang
force-pushed
the
multi_input_bugfix
branch
from
805b4a6 to
f7f3f67
Compare
|
To contributor, We have used 'Signed-off-by:' notation by default to handle the license issues, that result from contributors. Note that 'Is there a Signed-off-by line?' is important because lawyers tell us we must have to it to cleanly maintain the open-source license issues even though it has nothing to do with the code itself. |
There was a problem hiding this comment.
@EunjuYang, 💯 All CI checkers are successfully verified. Thanks.
|
Here is CI error log maybe TAOS bug Recognizing "-------------------" as the end of the message |
|
|
||
| ``` | ||
|
|
||
| - **[Note]** Users should feed the multi-input in reverse order because the model is structured in a reversed manner internally. This is a known issue for us, and we plan to address it soon. |
There was a problem hiding this comment.
I think it's great that it is illustrated in such a way that developers can easily understand it.
For this example app, no. But I worked on another model, named AuXFT, which takes multiple inputs; I compared the pytorch output for that example. For the case, the multi-input itself worked well. |
- This commit updates the model summary print of the layer with multiple
inputs.
[ASIS]
concat0 concat 1:1:14:2 input0
1:1:4:2 input1
1:1:8:2 input2
[TOBE]
concat0 concat 1:1:14:2 input0
input1
input2
Signed-off-by: Eunju Yang <[email protected]>
EunjuYang
force-pushed
the
multi_input_bugfix
branch
from
f7f3f67 to
06e5e6b
Compare
There was a problem hiding this comment.
@EunjuYang, 💯 All CI checkers are successfully verified. Thanks.
There was a problem hiding this comment.
It seems to be a good improvement. If the topological sort of the current implementation is always reversed in the order of the input layer, why don't we modify the implementation of the topological sort to arrange it to the regular arrangement? It seems that it can be used more intuitively without affecting performance.
Anyway, this PR seems to be a great example for users. LGTM!
[AS-IS]
[TO-BE]